R Notebook

library(tidyverse)

Warning message:
In `[<-.data.frame`(`*tmp*`, is_list, value = list(`23` = "<S3: blob>")) :
  replacement element 1 has 1 row to replace 0 rows

library(RSQLite)
library(dbplyr)
library(janitor)
library(lubridate)
library(datasets)
library(ggthemes)
library(gganimate)
library(modelr)
library(broom)
library(ggfortify)
library(infer)
library(MASS)
library(tseries)
library(forecast)
library(fable)
library(fabletools)
library(tsibble)
library(tsibbledata)
library(feasts)

1. Data Cleaning

Creating connection to the sqlite database and downloading fires dataset

# Connecting

conn <- dbConnect(SQLite(), "raw_data/FPA_FOD_20170508.sqlite")

# Pulling all the names of the tables in the database file

as.data.frame(dbListTables(conn))

# Making fires dataframe

fires <- tbl(conn, "Fires") %>% collect()

Seeing what other useful information is in the database. The majority are part of the database structure and are not readable in R.

# EPSG worldwide geodetic parameter dataset system
spatial_ref <- tbl(conn, "spatial_ref_sys_all") %>% collect()

# National Wildfire Coordinating Group unit abbreviations 
NWGG <- tbl(conn, "NWCG_UnitIDActive_20170109") %>% collect()

# Disconnect

dbDisconnect(conn)

Selecting columns of interest

fires_small <- fires %>%
  select(NWCG_REPORTING_AGENCY, SOURCE_REPORTING_UNIT_NAME, FIRE_NAME,
         FIRE_YEAR, DISCOVERY_DATE, DISCOVERY_DOY, DISCOVERY_TIME, CONT_DATE,
         CONT_DOY, CONT_TIME, STAT_CAUSE_CODE, STAT_CAUSE_DESCR, FIRE_SIZE, 
         FIRE_SIZE_CLASS, LATITUDE, LONGITUDE, OWNER_CODE, OWNER_DESCR, STATE, 
         COUNTY, FIPS_CODE, FIPS_NAME, Shape)

fires_small <- clean_names(fires_small)

Changing some columms to be factors

fires_small <- fires_small %>%
  mutate(nwcg_reporting_agency = as.factor(nwcg_reporting_agency)) %>%
  mutate(stat_cause_code = as.factor(stat_cause_code)) %>%
  mutate(fire_size_class = as.factor(fire_size_class)) %>%
  mutate(owner_descr = as.factor(owner_descr)) %>%
  mutate(state = as.factor(state)) 

Date is in Julian format, so overwriting with Gregorian format using year and day of year columns. Also adding in a ‘month of year column’ for future use.

fires_small <- fires_small %>%
  mutate(date_origin = as.Date(paste0(fire_year, "-01-01"))) %>%
  mutate(discovery_date = as.Date(discovery_doy, origin = date_origin)) %>%
  mutate(discovery_moy = month(discovery_date, label = TRUE)) %>%
  select(-date_origin)

2. Creating some initial visualisations

Fires per year

year_plot <- fires_small %>%
  group_by(fire_year) %>%
  summarise(num_fires =n())

`summarise()` ungrouping output (override with `.groups` argument)

year_plot %>%
  ggplot +
  aes(x = fire_year, y = num_fires) +
  geom_point() +
  ylim(0, 120000)

  # geom_col(fill = "dark blue", col ="white") +
  # geom_smooth(method = "lm", se = FALSE, colour = "red")

There is a lot of variation in the data between years. Visually it looks like a repeating pattern is occurring every 5 years or so with 4 peaks visible within this reporting period. Having looked at the historic weather for that date range these peaks seems to coincide with recorded heatwaves in 2000, 2006 and 2011.(1)

https://en.wikipedia.org/wiki/List_of_heat_waves

I will try to create a linear model to hopefully show any increase or decrease of fires over time. As this is time series data a linear model will not fit, but it could be interesting to see if it identifies a general underlying trend.

model <- lm(formula = num_fires ~ fire_year, data = year_plot)
summary(model)

Call:
lm(formula = num_fires ~ fire_year, data = year_plot)

Residuals:
   Min     1Q Median     3Q    Max 
-16835  -8688  -2049   9226  34793 

Coefficients:
             Estimate Std. Error t value Pr(>|t|)
(Intercept) -609543.8   756667.9  -0.806    0.429
fire_year       343.3      377.7   0.909    0.373

Residual standard error: 12810 on 22 degrees of freedom
Multiple R-squared:  0.03621,   Adjusted R-squared:  -0.007601 
F-statistic: 0.8265 on 1 and 22 DF,  p-value: 0.3731

tidy(model)

clean_names(glance(model))

The R Squared is quite low as expected from the widely spread plot and from the high p value we already know the model is not a good fit

autoplot(model)

`arrange_()` is deprecated as of dplyr 0.7.0.
Please use `arrange()` instead.
See vignette('programming') for more help
This warning is displayed once every 8 hours.
Call `lifecycle::last_warnings()` to see where this warning was generated.

The diagnostic plots agree that the model isn’t a great fit and there is likely to be a curve in the best fit line. As this is time series data we already know this is the case.

year_plot %>%
  add_predictions(model) %>%
  add_residuals(model)

year_plot

year_plot %>%
  ggplot(aes(x = fire_year)) +
  geom_point(aes(y = num_fires)) +
  geom_abline(
    intercept = model$coefficients[1],
    slope = model$coefficients[2],
    col = "red"
  ) +
  ylim(0, 120000)

NA

The plotted best fit line does show a slight increase, but as the P value is far too high I can not accept this model as a true representation of the occuring trend

Just to make sure I’m going to use bootstrapping using the same model

bootstrap_distribution_slope <- year_plot %>%
  specify(formula = num_fires ~ fire_year) %>%
  generate(reps = 10000, type = "bootstrap") %>%
  calculate(stat = "slope")

slope_ci95 <- bootstrap_distribution_slope %>%
  get_ci(level = 0.95, type = "percentile")
slope_ci95

bootstrap_distribution_slope %>%
  visualise(bins = 30) +
  shade_ci(endpoints = slope_ci95)

clean_names(tidy(model, conf.int = TRUE, conf.level = 0.95))

As 0 occurs within the 95% confidence intervals of -283 to +1075 it reinforces the fact that this model can not be used to explain if there are any positive or negative trends that are occurring in this data. It will be more use to use a model that is designed for time series and seasonal variations. For that I shall be also requiring more data points so I will now use monthly data and not yearly.

Fires per month

fires_small %>%
  mutate(year_month = make_date(fire_year, discovery_moy)) %>%
  group_by(year_month) %>%
  summarise(num_fires = n()) %>%
  ggplot +
  aes(x = year_month, y = num_fires) +
  geom_line(col = "dark blue")

`summarise()` ungrouping output (override with `.groups` argument)

Peaks are still shown to be occurring in the summers. The 2006 heatwave is especially visible.

Using the SARIMA model.

monthly <- fires_small %>%
  mutate(year_month = make_date(fire_year, discovery_moy)) %>%
  group_by(year_month) %>%
  summarise(num_fires = n())

`summarise()` ungrouping output (override with `.groups` argument)

write_csv(monthly, path = "clean_data/monthly.csv")

monthly

# Taking logs of data to smooth the volatility of the data

log_monthly <- log(monthly$num_fires)

# Autocorrelation plot to look to stationarity

acf(log_monthly)

pacf(log_monthly)

Definitely some repeating patterns in the lag plots which is likely to be some kind of seasonal variation.

adf.test(log_monthly)

p-value smaller than printed p-value

    Augmented Dickey-Fuller Test

data:  log_monthly
Dickey-Fuller = -8.3637, Lag order = 6, p-value = 0.01
alternative hypothesis: stationary

Agrees with a p-value of 0.01 that the hypothesis of having some kind of repeating pattern is correct and the null hypothesis of being stationary is not proven.

# Decomposing the time series data to examine any seasonal trends

arima_test <- ts(log_monthly, start = c(1992,01), frequency = 12)
components <- decompose(arima_test)
components

$x
          Jan      Feb      Mar      Apr      May      Jun      Jul      Aug
1992 8.143227 8.806724 9.127285 8.941415 9.070388 8.903136 8.915298 8.938138
1993 7.185387 8.472823 8.578288 8.821437 8.708144 8.843759 9.058703 8.960725
1994 8.090709 8.601534 9.350450 9.167537 8.837971 8.740497 9.369052 9.105979
1995 7.256297 8.708309 9.224342 9.209340 8.665613 8.504108 9.030855 8.870803
1996 8.081166 9.263976 9.172950 9.203718 8.804325 8.700847 9.174610 9.122274
1997 7.848934 8.214736 8.949495 9.236106 8.820995 8.372630 8.917445 8.714239
1998 7.087574 7.947679 8.777093 8.969542 8.466952 8.852522 9.247443 9.085684
1999 8.179760 8.692154 9.287579 9.276128 8.735686 8.714403 8.859363 9.422302
2000 8.549273 9.218507 9.363233 8.996776 8.994793 9.010669 9.375007 9.513034
2001 8.583543 8.814182 8.802673 9.169831 9.137662 8.722254 9.197356 9.007857
2002 8.183956 9.183791 9.147081 9.061028 8.915567 8.835065 9.274910 9.049115
2003 8.363342 7.975908 8.776321 9.198369 8.564458 8.573952 9.241839 9.116469
2004 8.196712 8.290794 9.458528 9.334326 8.620291 8.832296 9.060680 8.837246
2005 8.357728 8.513185 9.228082 9.450144 8.734077 8.720297 9.150697 9.050641
2006 9.169102 8.970305 9.847605 9.687009 9.056606 9.257701 9.657011 9.313799
2007 8.121183 8.983189 9.567665 9.258082 9.255027 9.016027 9.408781 9.171911
2008 8.640295 8.847360 9.209240 9.289706 8.885718 9.197356 9.233764 9.047586
2009 8.656955 9.178127 9.428431 9.303557 8.684739 8.652772 9.265775 8.819665
2010 7.827640 7.905442 9.117677 9.385134 8.539150 8.637107 9.113499 9.021719
2011 8.570355 9.216223 9.182249 9.056839 8.818482 9.156518 9.268704 9.343997
2012 8.318254 8.298540 8.993552 8.958797 8.674368 8.850661 9.395242 9.029777
2013 8.046229 8.091321 9.063347 8.907342 8.906935 8.632662 9.018090 9.022926
2014 8.679482 8.353497 9.121181 9.211739 8.752265 8.424200 9.058470 8.831128
2015 8.301770 8.594339 8.914223 9.157889 9.000483 8.745284 9.062188 9.056606
          Sep      Oct      Nov      Dec
1992 8.226573 8.208219 7.244228 6.993933
1993 8.497603 8.343316 8.087333 7.634337
1994 8.541300 7.914252 8.005701 7.271704
1995 8.883086 8.342602 7.999007 8.093157
1996 8.278936 8.055792 7.625595 7.380256
1997 8.689633 8.235095 7.368970 7.152269
1998 8.796944 8.534837 8.343316 8.052615
1999 9.148252 8.638348 8.919854 8.324336
2000 8.799963 8.885718 8.170186 7.811973
2001 8.494129 8.744010 9.289336 7.938802
2002 8.514189 7.579168 7.656337 7.486053
2003 8.423542 8.434898 8.074026 7.774015
2004 8.158516 7.653020 7.634337 7.584265
2005 8.927712 8.645059 8.981304 8.520587
2006 8.501064 8.346879 8.073091 8.130648
2007 8.798757 8.480114 8.353261 8.089176
2008 8.346168 8.417594 8.476163 7.855157
2009 8.443977 7.556428 7.990577 6.898715
2010 8.939843 9.066932 8.507951 8.363809
2011 8.996157 8.455743 8.209308 7.698936
2012 8.533067 8.206856 8.467162 7.656810
2013 8.472614 7.853993 8.240385 7.430707
2014 8.077137 8.028455 8.269757 7.580189
2015 8.625330 8.711608 8.105911 7.370860

$seasonal
             Jan         Feb         Mar         Apr         May         Jun
1992 -0.45855962 -0.02250662  0.50809392  0.53902883  0.15698768  0.12923426
1993 -0.45855962 -0.02250662  0.50809392  0.53902883  0.15698768  0.12923426
1994 -0.45855962 -0.02250662  0.50809392  0.53902883  0.15698768  0.12923426
1995 -0.45855962 -0.02250662  0.50809392  0.53902883  0.15698768  0.12923426
1996 -0.45855962 -0.02250662  0.50809392  0.53902883  0.15698768  0.12923426
1997 -0.45855962 -0.02250662  0.50809392  0.53902883  0.15698768  0.12923426
1998 -0.45855962 -0.02250662  0.50809392  0.53902883  0.15698768  0.12923426
1999 -0.45855962 -0.02250662  0.50809392  0.53902883  0.15698768  0.12923426
2000 -0.45855962 -0.02250662  0.50809392  0.53902883  0.15698768  0.12923426
2001 -0.45855962 -0.02250662  0.50809392  0.53902883  0.15698768  0.12923426
2002 -0.45855962 -0.02250662  0.50809392  0.53902883  0.15698768  0.12923426
2003 -0.45855962 -0.02250662  0.50809392  0.53902883  0.15698768  0.12923426
2004 -0.45855962 -0.02250662  0.50809392  0.53902883  0.15698768  0.12923426
2005 -0.45855962 -0.02250662  0.50809392  0.53902883  0.15698768  0.12923426
2006 -0.45855962 -0.02250662  0.50809392  0.53902883  0.15698768  0.12923426
2007 -0.45855962 -0.02250662  0.50809392  0.53902883  0.15698768  0.12923426
2008 -0.45855962 -0.02250662  0.50809392  0.53902883  0.15698768  0.12923426
2009 -0.45855962 -0.02250662  0.50809392  0.53902883  0.15698768  0.12923426
2010 -0.45855962 -0.02250662  0.50809392  0.53902883  0.15698768  0.12923426
2011 -0.45855962 -0.02250662  0.50809392  0.53902883  0.15698768  0.12923426
2012 -0.45855962 -0.02250662  0.50809392  0.53902883  0.15698768  0.12923426
2013 -0.45855962 -0.02250662  0.50809392  0.53902883  0.15698768  0.12923426
2014 -0.45855962 -0.02250662  0.50809392  0.53902883  0.15698768  0.12923426
2015 -0.45855962 -0.02250662  0.50809392  0.53902883  0.15698768  0.12923426
             Jul         Aug         Sep         Oct         Nov         Dec
1992  0.54027865  0.41448138 -0.05900239 -0.35734902 -0.47240336 -0.91828371
1993  0.54027865  0.41448138 -0.05900239 -0.35734902 -0.47240336 -0.91828371
1994  0.54027865  0.41448138 -0.05900239 -0.35734902 -0.47240336 -0.91828371
1995  0.54027865  0.41448138 -0.05900239 -0.35734902 -0.47240336 -0.91828371
1996  0.54027865  0.41448138 -0.05900239 -0.35734902 -0.47240336 -0.91828371
1997  0.54027865  0.41448138 -0.05900239 -0.35734902 -0.47240336 -0.91828371
1998  0.54027865  0.41448138 -0.05900239 -0.35734902 -0.47240336 -0.91828371
1999  0.54027865  0.41448138 -0.05900239 -0.35734902 -0.47240336 -0.91828371
2000  0.54027865  0.41448138 -0.05900239 -0.35734902 -0.47240336 -0.91828371
2001  0.54027865  0.41448138 -0.05900239 -0.35734902 -0.47240336 -0.91828371
2002  0.54027865  0.41448138 -0.05900239 -0.35734902 -0.47240336 -0.91828371
2003  0.54027865  0.41448138 -0.05900239 -0.35734902 -0.47240336 -0.91828371
2004  0.54027865  0.41448138 -0.05900239 -0.35734902 -0.47240336 -0.91828371
2005  0.54027865  0.41448138 -0.05900239 -0.35734902 -0.47240336 -0.91828371
2006  0.54027865  0.41448138 -0.05900239 -0.35734902 -0.47240336 -0.91828371
2007  0.54027865  0.41448138 -0.05900239 -0.35734902 -0.47240336 -0.91828371
2008  0.54027865  0.41448138 -0.05900239 -0.35734902 -0.47240336 -0.91828371
2009  0.54027865  0.41448138 -0.05900239 -0.35734902 -0.47240336 -0.91828371
2010  0.54027865  0.41448138 -0.05900239 -0.35734902 -0.47240336 -0.91828371
2011  0.54027865  0.41448138 -0.05900239 -0.35734902 -0.47240336 -0.91828371
2012  0.54027865  0.41448138 -0.05900239 -0.35734902 -0.47240336 -0.91828371
2013  0.54027865  0.41448138 -0.05900239 -0.35734902 -0.47240336 -0.91828371
2014  0.54027865  0.41448138 -0.05900239 -0.35734902 -0.47240336 -0.91828371
2015  0.54027865  0.41448138 -0.05900239 -0.35734902 -0.47240336 -0.91828371

$trend
          Jan      Feb      Mar      Apr      May      Jun      Jul      Aug
1992       NA       NA       NA       NA       NA       NA 8.419970 8.366148
1993 8.267328 8.274244 8.286478 8.303400 8.344158 8.405971 8.470376 8.513461
1994 8.627157 8.646141 8.654014 8.637957 8.616678 8.598167 8.548290 8.517972
1995 8.467242 8.443351 8.447793 8.479882 8.497451 8.531399 8.599996 8.657518
1996 8.709864 8.726332 8.711637 8.674514 8.647004 8.601741 8.562361 8.508966
1997 8.412648 8.384931 8.385042 8.409625 8.406404 8.386212 8.344989 8.302138
1998 8.278668 8.307894 8.327843 8.344803 8.397890 8.476002 8.559024 8.635552
1999 8.729376 8.727232 8.755895 8.774846 8.803181 8.838526 8.865244 8.902571
2000 8.975294 9.000560 8.989828 8.985623 8.964694 8.912109 8.892189 8.876770
2001 8.808100 8.779649 8.745857 8.727209 8.767936 8.819852 8.808487 8.807238
2002 8.836396 8.841347 8.843902 8.796202 8.679626 8.592719 8.581329 8.538475
2003 8.416299 8.417727 8.416757 8.448635 8.501694 8.531097 8.536152 8.542329
2004 8.642263 8.623080 8.600403 8.556782 8.505884 8.479657 8.478460 8.494435
2005 8.498049 8.510691 8.551632 8.625017 8.722476 8.817613 8.890433 8.943287
2006 9.126457 9.158518 9.151706 9.121505 9.071239 9.017149 8.957238 8.914112
2007 8.841629 8.825374 8.831866 8.849822 8.867047 8.876993 8.896894 8.912864
2008 8.857014 8.844541 8.820503 8.799040 8.801556 8.796926 8.787870 8.802346
2009 8.774752 8.766589 8.761167 8.729361 8.673246 8.613162 8.538755 8.451172
2010 8.359262 8.361336 8.390416 8.474015 8.558510 8.641113 8.733105 8.818667
2011 8.924335 8.944230 8.960005 8.936885 8.898975 8.858829 8.820622 8.771881
2012 8.677524 8.669704 8.637316 8.607650 8.608024 8.617012 8.603923 8.583954
2013 8.562348 8.546348 8.543543 8.526322 8.502170 8.483300 8.500265 8.537574
2014 8.550106 8.543797 8.519327 8.510118 8.518611 8.526063 8.516554 8.510851
2015 8.546749 8.556298 8.588535 8.639841 8.661479 8.645930       NA       NA
          Sep      Oct      Nov      Dec
1992 8.329360 8.301486 8.281394 8.263826
1993 8.550997 8.597592 8.617422 8.618529
1994 8.517166 8.513654 8.508214 8.491183
1995 8.678530 8.676154 8.681700 8.695677
1996 8.455937 8.447976 8.450020 8.437039
1997 8.283827 8.265537 8.239678 8.244922
1998 8.687842 8.721887 8.745858 8.751301
1999 8.927655 8.919168 8.918324 8.941465
2000 8.836566 8.820420 8.833584 8.827519
2001 8.836989 8.846805 8.833018 8.828464
2002 8.472699 8.462973 8.454066 8.428557
2003 8.583875 8.617965 8.625956 8.639047
2004 8.494100 8.489323 8.498890 8.498965
2005 8.988147 9.023830 9.047138 9.082969
2006 8.902985 8.873449 8.863844 8.862042
2007 8.892270 8.878654 8.864583 8.856751
2008 8.825261 8.834971 8.827174 8.796109
2009 8.385195 8.375646 8.372979 8.366260
2010 8.875974 8.864985 8.862945 8.896226
2011 8.725782 8.713834 8.703744 8.684995
2012 8.578228 8.578993 8.586539 8.587146
2013 8.550908 8.566001 8.572240 8.557109
2014 8.512263 8.501396 8.509494 8.533215
2015       NA       NA       NA       NA

$random
               Jan           Feb           Mar           Apr           May
1992            NA            NA            NA            NA            NA
1993 -0.6233809577  0.2210859405 -0.2162835416 -0.0209912924  0.2069981699
1994 -0.0778891050 -0.0220999711  0.1883422030 -0.0094486294  0.0643057584
1995 -0.7523847346  0.2874648696  0.2684548675  0.1904289896  0.0111744696
1996 -0.1701387014  0.5601507230 -0.0467807958 -0.0098239152  0.0003330778
1997 -0.1051545539 -0.1476887992  0.0563588626  0.2874512506  0.2576032032
1998 -0.7325343120 -0.3377092540 -0.0588436413  0.0857099361 -0.0879257640
1999 -0.0910557088 -0.0125705541  0.0235900729 -0.0377467128 -0.2244831524
2000  0.0325386191  0.2404537960 -0.1346890460 -0.5278762981 -0.1268889030
2001  0.2340021678  0.0570395770 -0.4512778124 -0.0964071698  0.2127385062
2002 -0.1938809368  0.3649509021 -0.2049145287 -0.2742032314  0.0789532160
2003  0.4056031774 -0.4193124026 -0.1485293225  0.2107048520 -0.0942237383
2004  0.0130089330 -0.3097795052  0.3500303370  0.2385150365 -0.0425804086
2005  0.3182391310  0.0250007048  0.1683557432  0.2860983870 -0.1453861909
2006  0.5012041268 -0.1657069193  0.1878044761  0.0264751592 -0.1716202969
2007 -0.2618864082  0.1803210569  0.2277049440 -0.1307680274  0.2309924287
2008  0.2418410992  0.0253251950 -0.1193573635 -0.0483632270 -0.0728262247
2009  0.3407631199  0.4340449334  0.1591698232  0.0351676969 -0.1454944827
2010 -0.0730631732 -0.4333881389  0.2191663716  0.3720896909 -0.1763473582
2011  0.1045792522  0.2944994489 -0.2858492688 -0.4190743966 -0.2374808430
2012  0.0992901576 -0.3486575784 -0.1518581830 -0.1878816359 -0.0906435908
2013 -0.0575592341 -0.4325199112  0.0117099464 -0.1580092312  0.2477773517
2014  0.5879358157 -0.1677931192  0.0937602986  0.1625925185  0.0766665120
2015  0.2135807753  0.0605475553 -0.1824058934 -0.0209812010  0.1820168093
               Jun           Jul           Aug           Sep           Oct
1992            NA -0.0449509633  0.1575087342 -0.0437845200  0.2640819290
1993  0.3085539972  0.0480481706  0.0327820440  0.0056074889  0.1030732247
1994  0.0130954944  0.2804833091  0.1735260613  0.0831362965 -0.2420523266
1995 -0.1565256653 -0.1094199513 -0.2011963255  0.2635580823  0.0237964250
1996 -0.0301283742  0.0719701407  0.1988264146 -0.1179987392 -0.0348344932
1997 -0.1428161289  0.0321772594 -0.0023803391  0.4648078450  0.3269074670
1998  0.2472854583  0.1481404093  0.0356503782  0.1681041910  0.1702988697
1999 -0.2533564798 -0.5461588727  0.1052488685  0.2795995836  0.0765297812
2000 -0.0306745492 -0.0574605672  0.2217828960  0.0223992451  0.4226463621
2001 -0.2268317961 -0.1514090624 -0.2138622447 -0.2838568677  0.2545536127
2002  0.1131112304  0.1533021643  0.0961580124  0.1004925057 -0.5264558037
2003 -0.0863793198  0.1654082837  0.1596583410 -0.1013309156  0.1742818811
2004  0.2234043597  0.0419409644 -0.0716710035 -0.2765810114 -0.4789540583
2005 -0.2265497504 -0.2800155713 -0.3071277015 -0.0014328522 -0.0214226481
2006  0.1113174957  0.1594943028 -0.0147943085 -0.3429184786 -0.1692202947
2007  0.0098004143 -0.0283915317 -0.1554343610 -0.0345114221 -0.0411904481
2008  0.2711958802 -0.0943844490 -0.1692410666 -0.4200907602 -0.0600280180
2009 -0.0896235184  0.1867412670 -0.0459878178  0.1177843980 -0.4618690336
2010 -0.1332398421 -0.1598845677 -0.2114296623  0.1228717981  0.5592954861
2011  0.1684543318 -0.0921967996  0.1576346081  0.3293774120  0.0992582136
2012  0.1044142300  0.2510404874  0.0313415740  0.0138405581 -0.0147870883
2013  0.0201278835 -0.0224538500  0.0708700702 -0.0192916141 -0.3546590289
2014 -0.2310973106  0.0016379769 -0.0942046225 -0.3761236738 -0.1155914613
2015 -0.0298794915            NA            NA            NA            NA
               Nov           Dec
1992 -0.5647630400 -0.3516096980
1993 -0.0576857033 -0.0659079015
1994 -0.0301097753 -0.3011952612
1995 -0.2102891499  0.3157635609
1996 -0.3520215661 -0.1384993286
1997 -0.3983046978 -0.1743696155
1998  0.0698607813  0.2195977677
1999  0.4739337289  0.3011555153
2000 -0.1909948168 -0.0972622783
2001  0.9287217703  0.0286215357
2002 -0.3253253092 -0.0242202677
2003 -0.0795267746  0.0532517319
2004 -0.3921497404  0.0035837222
2005  0.4065695347  0.3559022915
2006 -0.3183495899  0.1868897500
2007 -0.0389185632  0.1507084736
2008  0.1213922737 -0.0226681149
2009  0.0900014144 -0.5492616459
2010  0.1174087339  0.3858664418
2011 -0.0220324164 -0.0677754767
2012  0.3530267510 -0.0120520989
2013  0.1405486739 -0.2081184952
2014  0.2326660300 -0.0347420587
2015            NA            NA

$figure
 [1] -0.45855962 -0.02250662  0.50809392  0.53902883  0.15698768  0.12923426
 [7]  0.54027865  0.41448138 -0.05900239 -0.35734902 -0.47240336 -0.91828371

$type
[1] "additive"

attr(,"class")
[1] "decomposed.ts"

Definitely some seasonal trend in that data, but let’s have a look at a plot to make sure

plot(components)

The repeating seasonal trends can be seen very clearly in this plot.

# Using auto.arima to look at the best ARIMA configuration

fit_log_monthly <- auto.arima(arima_test, trace = TRUE, test = "kpss", ic = "bic")

 Fitting models using approximations to speed things up...

 ARIMA(2,0,2)(1,1,1)[12] with drift         : Inf
 ARIMA(0,0,0)(0,1,0)[12] with drift         : 386.3822
 ARIMA(1,0,0)(1,1,0)[12] with drift         : 210.0305
 ARIMA(0,0,1)(0,1,1)[12] with drift         : 206.1551
 ARIMA(0,0,0)(0,1,0)[12]                    : 380.8388
 ARIMA(0,0,1)(0,1,0)[12] with drift         : 332.9606
 ARIMA(0,0,1)(1,1,1)[12] with drift         : 199.5213
 ARIMA(0,0,1)(1,1,0)[12] with drift         : 244.7824
 ARIMA(0,0,1)(2,1,1)[12] with drift         : 204.6073
 ARIMA(0,0,1)(1,1,2)[12] with drift         : 194.5689
 ARIMA(0,0,1)(0,1,2)[12] with drift         : 211.4944
 ARIMA(0,0,1)(2,1,2)[12] with drift         : Inf
 ARIMA(0,0,0)(1,1,2)[12] with drift         : 249.2132
 ARIMA(1,0,1)(1,1,2)[12] with drift         : 149.1018
 ARIMA(1,0,1)(0,1,2)[12] with drift         : 161.5843
 ARIMA(1,0,1)(1,1,1)[12] with drift         : 146.0692
 ARIMA(1,0,1)(0,1,1)[12] with drift         : 157.8235
 ARIMA(1,0,1)(1,1,0)[12] with drift         : 197.9303
 ARIMA(1,0,1)(2,1,1)[12] with drift         : 161.8091
 ARIMA(1,0,1)(0,1,0)[12] with drift         : 292.8607
 ARIMA(1,0,1)(2,1,0)[12] with drift         : 188.0216
 ARIMA(1,0,1)(2,1,2)[12] with drift         : 153.8294
 ARIMA(1,0,0)(1,1,1)[12] with drift         : 153.6335
 ARIMA(2,0,1)(1,1,1)[12] with drift         : 147.4259
 ARIMA(1,0,2)(1,1,1)[12] with drift         : 151.6193
 ARIMA(0,0,0)(1,1,1)[12] with drift         : 256.4293
 ARIMA(0,0,2)(1,1,1)[12] with drift         : 183.564
 ARIMA(2,0,0)(1,1,1)[12] with drift         : 141.9116
 ARIMA(2,0,0)(0,1,1)[12] with drift         : 160.3183
 ARIMA(2,0,0)(1,1,0)[12] with drift         : 201.1907
 ARIMA(2,0,0)(2,1,1)[12] with drift         : 165.2805
 ARIMA(2,0,0)(1,1,2)[12] with drift         : Inf
 ARIMA(2,0,0)(0,1,0)[12] with drift         : 294.3017
 ARIMA(2,0,0)(0,1,2)[12] with drift         : 163.7419
 ARIMA(2,0,0)(2,1,0)[12] with drift         : 189.6952
 ARIMA(2,0,0)(2,1,2)[12] with drift         : 160.4628
 ARIMA(3,0,0)(1,1,1)[12] with drift         : 140.895
 ARIMA(3,0,0)(0,1,1)[12] with drift         : 166.2635
 ARIMA(3,0,0)(1,1,0)[12] with drift         : 202.3331
 ARIMA(3,0,0)(2,1,1)[12] with drift         : 167.8427
 ARIMA(3,0,0)(1,1,2)[12] with drift         : 146.502
 ARIMA(3,0,0)(0,1,0)[12] with drift         : 299.1173
 ARIMA(3,0,0)(0,1,2)[12] with drift         : 169.2991
 ARIMA(3,0,0)(2,1,0)[12] with drift         : 190.1357
 ARIMA(3,0,0)(2,1,2)[12] with drift         : 164.0628
 ARIMA(4,0,0)(1,1,1)[12] with drift         : 148.044
 ARIMA(3,0,1)(1,1,1)[12] with drift         : Inf
 ARIMA(4,0,1)(1,1,1)[12] with drift         : 152.986
 ARIMA(3,0,0)(1,1,1)[12]                    : 135.3069
 ARIMA(3,0,0)(0,1,1)[12]                    : 161.3958
 ARIMA(3,0,0)(1,1,0)[12]                    : 196.7164
 ARIMA(3,0,0)(2,1,1)[12]                    : 162.2665
 ARIMA(3,0,0)(1,1,2)[12]                    : 140.8981
 ARIMA(3,0,0)(0,1,0)[12]                    : 293.5803
 ARIMA(3,0,0)(0,1,2)[12]                    : 164.2536
 ARIMA(3,0,0)(2,1,0)[12]                    : 184.517
 ARIMA(3,0,0)(2,1,2)[12]                    : 158.4784
 ARIMA(2,0,0)(1,1,1)[12]                    : 136.5386
 ARIMA(4,0,0)(1,1,1)[12]                    : 142.4368
 ARIMA(3,0,1)(1,1,1)[12]                    : 140.8894
 ARIMA(2,0,1)(1,1,1)[12]                    : 142.018
 ARIMA(4,0,1)(1,1,1)[12]                    : 147.3867

 Now re-fitting the best model(s) without approximations...

 ARIMA(3,0,0)(1,1,1)[12]                    : 129.5424

 Best model: ARIMA(3,0,0)(1,1,1)[12]                    

fit_log_monthly

Series: arima_test 
ARIMA(3,0,0)(1,1,1)[12] 

Coefficients:
         ar1     ar2     ar3     sar1     sma1
      0.4104  0.1781  0.1091  -0.1107  -0.8207
s.e.  0.0600  0.0644  0.0610   0.0734   0.0515

sigma^2 estimated as 0.07962:  log likelihood=-47.91
AIC=107.82   AICc=108.13   BIC=129.54

auto.arima has chosen the best model with the lowest BIC.

confint(fit_log_monthly)

           2.5 %      97.5 %
ar1   0.29277203  0.52793840
ar2   0.05192296  0.30419421
ar3  -0.01051291  0.22875135
sar1 -0.25453068  0.03310203
sma1 -0.92163837 -0.71985077

Non-log data

monthly_num_fires <- monthly$num_fires

acf(monthly_num_fires)

pacf(monthly_num_fires)

adf.test(monthly_num_fires)

p-value smaller than printed p-value

    Augmented Dickey-Fuller Test

data:  monthly_num_fires
Dickey-Fuller = -7.9715, Lag order = 6, p-value = 0.01
alternative hypothesis: stationary

monthly_arima <- ts(monthly_num_fires, start = c(1991, 01), frequency = 12)

components_monthly <- decompose(monthly_arima)
components_monthly

$x
       Jan   Feb   Mar   Apr   May   Jun   Jul   Aug   Sep   Oct   Nov   Dec
1991  3440  6679  9203  7642  8694  7355  7445  7617  3739  3671  1400  1090
1992  1320  4783  5315  6778  6052  6931  8593  7791  4903  4202  3253  2068
1993  3264  5440 11504  9581  6891  6251 11720  9009  5122  2736  2998  1439
1994  1417  6053 10141  9990  5800  4935  8357  7121  7209  4199  2978  3272
1995  3233 10551  9633  9934  6663  6008  9649  9157  3940  3152  2050  1604
1996  2563  3695  7704 10261  6775  4327  7461  6089  5941  3771  1586  1277
1997  1197  2829  6484  7860  4755  6992 10378  8828  6614  5089  4202  3142
1998  3568  5956 10803 10680  6221  6090  7040 12361  9398  5644  7479  4123
1999  5163 10082 11652  8077  8061  8190 11790 13535  6634  7228  3534  2470
2000  5343  6729  6652  9603  9299  6138  9871  8167  4886  6273 10822  2804
2001  3583  9738  9387  8613  7447  6871 10667  8511  4985  1957  2114  1783
2002  4287  2910  6479  9881  5242  5292 10320  9104  4553  4605  3210  2378
2003  3629  3987 12817 11320  5543  6852  8610  6886  3493  2107  2068  1967
2004  4263  4980 10179 12710  6211  6126  9421  8524  7538  5682  7953  5017
2005  9596  7866 18913 16107  8575 10485 15631 11090  4920  4217  3207  3397
2006  3365  7968 14295 10489 10457  8234 12195  9623  6626  4818  4244  3259
2007  5655  6956  9989 10826  7228  9871 10237  8498  4214  4526  4799  2579
2008  5750  9683 12437 10977  5912  5726 10570  6766  4647  1913  2953   991
2009  2509  2712  9115 11910  5111  5637  9077  8281  7630  8664  4954  4289
2010  5273 10059  9723  8577  6758  9476 10601 11430  8072  4702  3675  2206
2011  4098  4018  8051  7776  5851  6979 12031  8348  5080  3666  4756  2115
2012  3122  3266  8633  7386  7383  5612  8251  8291  4782  2576  3791  1687
2013  5881  4245  9147 10014  6325  4556  8591  6844  3220  3067  3904  1959
2014  4031  5401  7437  9489  8107  6281  8623  8575  5571  6073  3314  1589

$seasonal
            Jan        Feb        Mar        Apr        May        Jun
1991 -2522.1208  -447.8599  3311.5640  3406.0242   260.2923   133.8774
1992 -2522.1208  -447.8599  3311.5640  3406.0242   260.2923   133.8774
1993 -2522.1208  -447.8599  3311.5640  3406.0242   260.2923   133.8774
1994 -2522.1208  -447.8599  3311.5640  3406.0242   260.2923   133.8774
1995 -2522.1208  -447.8599  3311.5640  3406.0242   260.2923   133.8774
1996 -2522.1208  -447.8599  3311.5640  3406.0242   260.2923   133.8774
1997 -2522.1208  -447.8599  3311.5640  3406.0242   260.2923   133.8774
1998 -2522.1208  -447.8599  3311.5640  3406.0242   260.2923   133.8774
1999 -2522.1208  -447.8599  3311.5640  3406.0242   260.2923   133.8774
2000 -2522.1208  -447.8599  3311.5640  3406.0242   260.2923   133.8774
2001 -2522.1208  -447.8599  3311.5640  3406.0242   260.2923   133.8774
2002 -2522.1208  -447.8599  3311.5640  3406.0242   260.2923   133.8774
2003 -2522.1208  -447.8599  3311.5640  3406.0242   260.2923   133.8774
2004 -2522.1208  -447.8599  3311.5640  3406.0242   260.2923   133.8774
2005 -2522.1208  -447.8599  3311.5640  3406.0242   260.2923   133.8774
2006 -2522.1208  -447.8599  3311.5640  3406.0242   260.2923   133.8774
2007 -2522.1208  -447.8599  3311.5640  3406.0242   260.2923   133.8774
2008 -2522.1208  -447.8599  3311.5640  3406.0242   260.2923   133.8774
2009 -2522.1208  -447.8599  3311.5640  3406.0242   260.2923   133.8774
2010 -2522.1208  -447.8599  3311.5640  3406.0242   260.2923   133.8774
2011 -2522.1208  -447.8599  3311.5640  3406.0242   260.2923   133.8774
2012 -2522.1208  -447.8599  3311.5640  3406.0242   260.2923   133.8774
2013 -2522.1208  -447.8599  3311.5640  3406.0242   260.2923   133.8774
2014 -2522.1208  -447.8599  3311.5640  3406.0242   260.2923   133.8774
            Jul        Aug        Sep        Oct        Nov        Dec
1991  3400.9897  2244.1908  -955.7295 -2246.3545 -2532.7675 -4052.1063
1992  3400.9897  2244.1908  -955.7295 -2246.3545 -2532.7675 -4052.1063
1993  3400.9897  2244.1908  -955.7295 -2246.3545 -2532.7675 -4052.1063
1994  3400.9897  2244.1908  -955.7295 -2246.3545 -2532.7675 -4052.1063
1995  3400.9897  2244.1908  -955.7295 -2246.3545 -2532.7675 -4052.1063
1996  3400.9897  2244.1908  -955.7295 -2246.3545 -2532.7675 -4052.1063
1997  3400.9897  2244.1908  -955.7295 -2246.3545 -2532.7675 -4052.1063
1998  3400.9897  2244.1908  -955.7295 -2246.3545 -2532.7675 -4052.1063
1999  3400.9897  2244.1908  -955.7295 -2246.3545 -2532.7675 -4052.1063
2000  3400.9897  2244.1908  -955.7295 -2246.3545 -2532.7675 -4052.1063
2001  3400.9897  2244.1908  -955.7295 -2246.3545 -2532.7675 -4052.1063
2002  3400.9897  2244.1908  -955.7295 -2246.3545 -2532.7675 -4052.1063
2003  3400.9897  2244.1908  -955.7295 -2246.3545 -2532.7675 -4052.1063
2004  3400.9897  2244.1908  -955.7295 -2246.3545 -2532.7675 -4052.1063
2005  3400.9897  2244.1908  -955.7295 -2246.3545 -2532.7675 -4052.1063
2006  3400.9897  2244.1908  -955.7295 -2246.3545 -2532.7675 -4052.1063
2007  3400.9897  2244.1908  -955.7295 -2246.3545 -2532.7675 -4052.1063
2008  3400.9897  2244.1908  -955.7295 -2246.3545 -2532.7675 -4052.1063
2009  3400.9897  2244.1908  -955.7295 -2246.3545 -2532.7675 -4052.1063
2010  3400.9897  2244.1908  -955.7295 -2246.3545 -2532.7675 -4052.1063
2011  3400.9897  2244.1908  -955.7295 -2246.3545 -2532.7675 -4052.1063
2012  3400.9897  2244.1908  -955.7295 -2246.3545 -2532.7675 -4052.1063
2013  3400.9897  2244.1908  -955.7295 -2246.3545 -2532.7675 -4052.1063
2014  3400.9897  2244.1908  -955.7295 -2246.3545 -2532.7675 -4052.1063

$trend
           Jan       Feb       Mar       Apr       May       Jun       Jul
1991        NA        NA        NA        NA        NA        NA  5576.250
1992  4726.250  4781.333  4837.083  4907.708  5007.042  5125.000  5246.750
1993  6275.375  6456.417  6516.292  6464.333  6392.625  6355.792  6252.625
1994  5806.542  5587.750  5596.042  5743.958  5804.083  5879.625  6031.667
1995  6650.333  6789.000  6737.625  6557.792  6475.500  6367.333  6269.917
1996  5315.250  5096.250  5051.792  5160.958  5167.417  5134.458  5063.917
1997  4808.375  5044.042  5186.208  5269.167  5433.083  5619.792  5796.292
1998  6658.500  6666.625  6929.833  7068.958  7228.625  7406.042  7513.375
1999  8303.750  8550.583  8484.333  8435.167  8336.792  8103.542  8042.167
2000  7332.958  7029.333  6732.833  6620.208  6884.083  7201.667  7142.250
2001  7405.000  7452.500  7470.958  7295.250  6752.583  6347.208  6334.000
2002  5327.875  5338.125  5344.833  5437.167  5593.167  5663.625  5661.000
2003  6455.250  6291.583  6155.000  6006.750  5855.083  5790.375  5799.667
2004  5833.792  5935.833  6172.625  6490.125  6884.292  7256.583  7605.875
2005  9898.500 10264.167 10262.000 10091.875  9833.083  9567.833  9240.708
2006  7962.667  7758.375  7768.333  7864.458  7932.708  7970.167  8059.833
2007  7525.917  7397.458  7250.083  7137.417  7148.375  7143.167  7118.792
2008  7125.375  7067.083  7012.958  6922.125  6736.333  6593.250  6392.042
2009  5340.625  5341.542  5528.958  5934.542  6299.208  6520.000  6772.583
2010  7793.583  7988.292  8137.917  7991.250  7772.875  7632.792  7497.042
2011  6514.500  6445.667  6192.583  6024.750  6026.625  6067.875  6023.417
2012  5792.333  5632.458  5617.667  5559.833  5474.208  5416.167  5513.292
2013  5809.667  5763.542  5638.167  5593.542  5618.708  5634.750  5569.000
2014  5695.583  5769.042  5939.125  6162.333  6263.000  6223.000        NA
           Aug       Sep       Oct       Nov       Dec
1991  5408.917  5167.917  4969.917  4823.833  4696.083
1992  5355.125  5640.375  6015.042  6166.792  6173.417
1993  6201.208  6169.958  6130.208  6101.792  6001.500
1994  6294.750  6461.000  6437.500  6471.125  6551.792
1995  5956.333  5590.292  5523.542  5541.833  5476.458
1996  4970.917  4884.000  4733.125  4548.917  4575.792
1997  6025.375  6335.625  6633.083  6811.667  6835.167
1998  7751.750  7959.042  7885.958  7854.167  8018.333
1999  7909.958  7561.917  7417.167  7532.333  7498.417
2000  7194.292  7433.625  7506.333  7387.917  7341.292
2001  6078.833  5673.167  5604.833  5565.792  5408.125
2002  5678.458  5987.417  6311.458  6383.958  6461.500
2003  5867.458  5798.917  5746.917  5832.667  5830.250
2004  7948.333  8432.500  8937.958  9178.000  9458.125
2005  8985.333  8797.167  8370.667  8215.000  8199.625
2006  8113.083  7891.500  7726.125  7605.625  7539.292
2007  7236.375  7452.000  7560.292  7511.750  7284.208
2008  5966.542  5537.667  5438.125  5443.625  5406.542
2009  7193.875  7525.333  7411.792  7341.542  7570.125
2010  7196.375  6875.000  6771.958  6700.792  6558.958
2011  5951.417  5944.333  5952.333  5999.917  6006.792
2012  5669.042  5731.250  5862.167  5927.583  5839.500
2013  5540.083  5517.000  5423.875  5476.250  5622.375
2014        NA        NA        NA        NA        NA

$random
              Jan          Feb          Mar          Apr          May
1991           NA           NA           NA           NA           NA
1992  -884.129227   449.526570 -2833.647343 -1535.732488   784.666063
1993  -489.254227  -568.556763  1676.144324  -289.357488   238.082729
1994 -1867.420894   913.109903  1233.394324   840.017512  -264.375604
1995  -895.212560  4209.859903  -416.189010   -29.815821   -72.792271
1996  -230.129227  -953.390097  -659.355676  1694.017512  1347.291063
1997 -1089.254227 -1767.181763 -2013.772343  -815.190821  -938.375604
1998  -568.379227  -262.765097   561.602657   205.017512 -1267.917271
1999  -618.629227  1979.276570  -143.897343 -3764.190821  -536.083937
2000   532.162440   147.526570 -3392.397343  -423.232488  2154.624396
2001 -1299.879227  2733.359903 -1395.522343 -2088.274155   434.124396
2002  1481.245773 -1980.265097 -2177.397343  1037.809179  -611.458937
2003  -304.129227 -1856.723430  3350.435990  1907.225845  -572.375604
2004   951.329106  -507.973430   694.810990  2813.850845  -933.583937
2005  2219.620773 -1950.306763  5339.435990  2609.100845 -1518.375604
2006 -2075.545894   657.484903  3215.102657  -781.482488  2263.999396
2007   651.204106     6.401570  -572.647343   282.559179  -180.667271
2008  1146.745773  3063.776570  2112.477657   648.850845 -1084.625604
2009  -309.504227 -2181.681763   274.477657  2569.434179 -1448.500604
2010     1.537440  2518.568237 -1726.480676 -2820.274155 -1275.167271
2011   105.620773 -1979.806763 -1453.147343 -1654.774155  -435.917271
2012  -148.212560 -1918.598430  -296.230676 -1579.857488  1648.499396
2013  2593.454106 -1070.681763   197.269324  1014.434179   445.999396
2014   857.537440    79.818237 -1813.689010   -79.357488  1583.707729
              Jun          Jul          Aug          Sep          Oct
1991           NA -1532.239734   -36.107488  -473.187198   947.437802
1992  1672.122585   -54.739734   191.684179   218.354469   433.312802
1993  -238.669082  2066.385266   563.600845   -92.228865 -1147.853865
1994 -1078.502415 -1075.656401 -1417.940821  1703.729469     7.854469
1995  -493.210749   -21.906401   956.475845  -694.562198  -125.187198
1996  -941.335749 -1003.906401 -1126.107488  2012.729469  1284.229469
1997  1238.330918  1180.718599   558.434179  1234.104469   702.271135
1998 -1449.919082 -3874.364734  2365.059179  2394.687802     4.396135
1999   -47.419082   346.843599  3380.850845    27.812802  2057.187802
2000 -1197.544082  -672.239734 -1271.482488 -1591.895531  1013.021135
2001   389.914251   932.010266   187.975845   267.562802 -1401.478865
2002  -505.502415  1258.010266  1181.350845  -478.687198   539.896135
2003   927.747585  -590.656401 -1225.649155 -1350.187198 -1393.562198
2004 -1264.460749 -1585.864734 -1668.524155    61.229469 -1009.603865
2005   783.289251  2989.301932  -139.524155 -2921.437198 -1907.312198
2006   129.955918   734.176932  -734.274155  -309.770531  -661.770531
2007  2593.955918  -282.781401  -982.565821 -2282.270531  -787.937198
2008 -1001.127415   776.968599 -1444.732488    65.062802 -1278.770531
2009 -1016.877415 -1096.573068 -1157.065821  1060.396135  3498.562802
2010  1709.330918  -297.031401  1989.434179  2152.729469   176.396135
2011   777.247585  2606.593599   152.392512    91.396135   -39.978865
2012    61.955918  -663.281401   377.767512     6.479469 -1039.812198
2013 -1212.627415  -378.989734  -940.274155 -1341.270531  -110.520531
2014   -75.877415           NA           NA           NA           NA
              Nov          Dec
1991  -891.065821   446.022947
1992  -381.024155   -53.310386
1993  -571.024155  -510.393720
1994  -960.357488   772.314614
1995  -959.065821   179.647947
1996  -430.149155   753.314614
1997   -76.899155   358.939614
1998  2157.600845   156.772947
1999 -1465.565821  -976.310386
2000  5966.850845  -485.185386
2001  -919.024155   426.981280
2002  -641.190821   -31.393720
2003 -1231.899155   188.856280
2004  1307.767512  -389.018720
2005 -2475.232488  -750.518720
2006  -828.857488  -228.185386
2007  -179.982488  -653.102053
2008    42.142512  -363.435386
2009   145.225845   770.981280
2010  -493.024155  -300.852053
2011  1288.850845   160.314614
2012   396.184179  -100.393720
2013   960.517512   388.731280
2014           NA           NA

$figure
 [1] -2522.1208  -447.8599  3311.5640  3406.0242   260.2923   133.8774
 [7]  3400.9897  2244.1908  -955.7295 -2246.3545 -2532.7675 -4052.1063

$type
[1] "additive"

attr(,"class")
[1] "decomposed.ts"

plot(components_monthly)

fit_monthly <- auto.arima(monthly_arima, trace = TRUE, test = "kpss", ic = "bic")

 Fitting models using approximations to speed things up...

 ARIMA(2,0,2)(1,1,1)[12] with drift         : Inf
 ARIMA(0,0,0)(0,1,0)[12] with drift         : 4940.673
 ARIMA(1,0,0)(1,1,0)[12] with drift         : 4820.157
 ARIMA(0,0,1)(0,1,1)[12] with drift         : 4780.66
 ARIMA(0,0,0)(0,1,0)[12]                    : 4935.076
 ARIMA(0,0,1)(0,1,0)[12] with drift         : 4905.171
 ARIMA(0,0,1)(1,1,1)[12] with drift         : 4792.821
 ARIMA(0,0,1)(0,1,2)[12] with drift         : 4785.781
 ARIMA(0,0,1)(1,1,0)[12] with drift         : 4835.391
 ARIMA(0,0,1)(1,1,2)[12] with drift         : 4789.521
 ARIMA(0,0,0)(0,1,1)[12] with drift         : 4823.973
 ARIMA(1,0,1)(0,1,1)[12] with drift         : 4757.861
 ARIMA(1,0,1)(0,1,0)[12] with drift         : 4893.217
 ARIMA(1,0,1)(1,1,1)[12] with drift         : 4762.762
 ARIMA(1,0,1)(0,1,2)[12] with drift         : 4763.454
 ARIMA(1,0,1)(1,1,0)[12] with drift         : 4810.19
 ARIMA(1,0,1)(1,1,2)[12] with drift         : 4766.262
 ARIMA(1,0,0)(0,1,1)[12] with drift         : 4766.611
 ARIMA(2,0,1)(0,1,1)[12] with drift         : 4764.112
 ARIMA(1,0,2)(0,1,1)[12] with drift         : 4762.623
 ARIMA(0,0,2)(0,1,1)[12] with drift         : 4780.193
 ARIMA(2,0,0)(0,1,1)[12] with drift         : 4767.588
 ARIMA(2,0,2)(0,1,1)[12] with drift         : 4767.744
 ARIMA(1,0,1)(0,1,1)[12]                    : 4753.056
 ARIMA(1,0,1)(0,1,0)[12]                    : 4887.636
 ARIMA(1,0,1)(1,1,1)[12]                    : 4757.169
 ARIMA(1,0,1)(0,1,2)[12]                    : 4758.612
 ARIMA(1,0,1)(1,1,0)[12]                    : 4804.576
 ARIMA(1,0,1)(1,1,2)[12]                    : 4760.879
 ARIMA(0,0,1)(0,1,1)[12]                    : 4775.902
 ARIMA(1,0,0)(0,1,1)[12]                    : 4761.7
 ARIMA(2,0,1)(0,1,1)[12]                    : 4759.357
 ARIMA(1,0,2)(0,1,1)[12]                    : 4757.868
 ARIMA(0,0,0)(0,1,1)[12]                    : 4819.725
 ARIMA(0,0,2)(0,1,1)[12]                    : 4775.225
 ARIMA(2,0,0)(0,1,1)[12]                    : 4762.753
 ARIMA(2,0,2)(0,1,1)[12]                    : 4763.001

 Now re-fitting the best model(s) without approximations...

 ARIMA(1,0,1)(0,1,1)[12]                    : Inf
 ARIMA(1,0,1)(1,1,1)[12]                    : Inf
 ARIMA(1,0,1)(0,1,1)[12] with drift         : Inf
 ARIMA(1,0,2)(0,1,1)[12]                    : Inf
 ARIMA(1,0,1)(0,1,2)[12]                    : Inf
 ARIMA(2,0,1)(0,1,1)[12]                    : Inf
 ARIMA(1,0,1)(1,1,2)[12]                    : Inf
 ARIMA(1,0,0)(0,1,1)[12]                    : Inf
 ARIMA(1,0,2)(0,1,1)[12] with drift         : Inf
 ARIMA(2,0,0)(0,1,1)[12]                    : Inf
 ARIMA(1,0,1)(1,1,1)[12] with drift         : Inf
 ARIMA(2,0,2)(0,1,1)[12]                    : Inf
 ARIMA(1,0,1)(0,1,2)[12] with drift         : Inf
 ARIMA(2,0,1)(0,1,1)[12] with drift         : Inf
 ARIMA(1,0,1)(1,1,2)[12] with drift         : Inf
 ARIMA(1,0,0)(0,1,1)[12] with drift         : Inf
 ARIMA(2,0,0)(0,1,1)[12] with drift         : Inf
 ARIMA(2,0,2)(0,1,1)[12] with drift         : Inf
 ARIMA(0,0,2)(0,1,1)[12]                    : Inf
 ARIMA(0,0,1)(0,1,1)[12]                    : Inf
 ARIMA(0,0,2)(0,1,1)[12] with drift         : Inf
 ARIMA(0,0,1)(0,1,1)[12] with drift         : Inf
 ARIMA(0,0,1)(0,1,2)[12] with drift         : Inf
 ARIMA(0,0,1)(1,1,2)[12] with drift         : Inf
 ARIMA(0,0,1)(1,1,1)[12] with drift         : Inf
 ARIMA(1,0,1)(1,1,0)[12]                    : 4988.73

 Best model: ARIMA(1,0,1)(1,1,0)[12]                    

fit_monthly

Series: monthly_arima 
ARIMA(1,0,1)(1,1,0)[12] 

Coefficients:
         ar1      ma1     sar1
      0.8466  -0.5434  -0.5353
s.e.  0.0564   0.0883   0.0497

sigma^2 estimated as 3802460:  log likelihood=-2483.12
AIC=4974.25   AICc=4974.4   BIC=4988.73

confint(fit_monthly)

          2.5 %     97.5 %
ar1   0.7361480  0.9571085
ma1  -0.7164356 -0.3703286
sar1 -0.6326515 -0.4379326

Fires per day

fires_small %>%
  group_by(discovery_date) %>%
  summarise(num_fires = n()) %>%
  ggplot +
  aes(x = discovery_date, y = num_fires) +
  geom_line(col = "dark blue") 

This shows a typical time series plot with a cyclic variation due to warmer weather in the summer time.

Fires by day of year

fires_small %>%
  group_by(discovery_doy) %>%
  summarise(num_fires = n()) %>%
  ggplot(aes(x = discovery_doy, y = num_fires)) +
  geom_line(col = "dark blue")

The are peaks around day 60-110 and a big peak around 180.

Checking the data to see where the peak occurs

fires_small %>%
  group_by(discovery_doy) %>%
  summarise(num_fires = n()) %>%
  arrange(desc(num_fires))

`summarise()` ungrouping output (override with `.groups` argument)

The 2 highest days of the year are on 185 and 186, which happens to be Independence Day (4th July) on a normal year and a leap year retrospectively. So I imagine most of the extra fires (literally over double the normal amount) are caused by fireworks.

Fires by month of year

fires_small %>%
  group_by(discovery_moy) %>%
  summarise(num_fires = n()) %>%
  ggplot(aes(x = discovery_moy, y = num_fires)) +
  geom_col(fill = "dark blue", col = "white")

`summarise()` ungrouping output (override with `.groups` argument)

There are 2 definite peaks during the year. March and April are probably due to the US “Spring Break”, where schools and Universities are stopped and so families are likely to be on vacation during that period possibly visiting National Parks. July and August is also Summer Break for school with both families visiting Parks and hot weather likely causes of fire outbreaks.

Fires by cause

options(scipen = 999)

fires_small %>%
  group_by(stat_cause_descr) %>%
  summarise(num_fires = n()) %>%
  ggplot +
  aes(reorder(x = stat_cause_descr, num_fires), y = num_fires) +
  geom_col(fill = "dark blue") + 
  coord_flip() 

`summarise()` ungrouping output (override with `.groups` argument)

Fire avg size by cause

fires_small %>%
  group_by(stat_cause_descr) %>%
  summarise(avg_size = mean(fire_size)) %>%
  ggplot +
  aes(reorder(x = stat_cause_descr, avg_size), y = avg_size) +
  geom_col(fill = "dark blue") + 
  coord_flip()

`summarise()` ungrouping output (override with `.groups` argument)

Avg burn time by cause

fires_small %>%
  summarise(num_na = sum(is.na(cont_date)))

Literally half the data is missing for burn time, making it very difficult to do any meaningful analysis

Fires by size

fires_small %>%
  group_by(fire_size_class) %>%
  summarise(num_fires = n()) %>%
  ggplot +
  aes(x = fire_size_class, y = num_fires, fill = fire_size_class) +
  geom_col() +
  scale_fill_manual(values = c("red", "orange", "yellow", "green", "blue", 
                               "purple", "black"),
                    name = "Fire Size Classification",
                    breaks = c("A", "B", "C", "D", "E", "F", "G"),
                    labels = c("A: < 1/4 acre", "B: 1/4 to 10 acres", "C: 10 to 100 acres",
                               "D: 100 to 300 acres", "E: 300 to 1000 acres",
                               "F: 1000 to 5000 acres", "G: More than 5000 acres"))

`summarise()` ungrouping output (override with `.groups` argument)

Geo Spatial wrangling

To make it easier to visually detect frequency of wildfires between states I want display it in a map format. As I’m using ggplot2 already I’m going to also use it for maps with the geom_polygon(), coord_map() along with the ggthemes theme_map() functions.

I’m not entirely sure what geo-spatial information is being held with in the sqlite database file, I’ve made a few attempts to retrieve it but have been unsuccessful. Therefore I’m going to utelise the datasets package which includes various bits of information on the US States, including coordinates for state boundaries.

# State boundary co-ordinates from 'datasets' package

state_map <- map_data("state")
state_map

Annoyingly it doesn’t have the abbreviation of the State, only the full name so I need to add that in. Luckily the ‘datasets’ package also has a vector of States names and abbreviations so I shall make a tibble with them both in.

state.abb

 [1] "AL" "AK" "AZ" "AR" "CA" "CO" "CT" "DE" "FL" "GA" "HI" "ID" "IL" "IN" "IA"
[16] "KS" "KY" "LA" "ME" "MD" "MA" "MI" "MN" "MS" "MO" "MT" "NE" "NV" "NH" "NJ"
[31] "NM" "NY" "NC" "ND" "OH" "OK" "OR" "PA" "RI" "SC" "SD" "TN" "TX" "UT" "VT"
[46] "VA" "WA" "WV" "WI" "WY"

state.name

 [1] "Alabama"        "Alaska"         "Arizona"        "Arkansas"      
 [5] "California"     "Colorado"       "Connecticut"    "Delaware"      
 [9] "Florida"        "Georgia"        "Hawaii"         "Idaho"         
[13] "Illinois"       "Indiana"        "Iowa"           "Kansas"        
[17] "Kentucky"       "Louisiana"      "Maine"          "Maryland"      
[21] "Massachusetts"  "Michigan"       "Minnesota"      "Mississippi"   
[25] "Missouri"       "Montana"        "Nebraska"       "Nevada"        
[29] "New Hampshire"  "New Jersey"     "New Mexico"     "New York"      
[33] "North Carolina" "North Dakota"   "Ohio"           "Oklahoma"      
[37] "Oregon"         "Pennsylvania"   "Rhode Island"   "South Carolina"
[41] "South Dakota"   "Tennessee"      "Texas"          "Utah"          
[45] "Vermont"        "Virginia"       "Washington"     "West Virginia" 
[49] "Wisconsin"      "Wyoming"       

state_list <- tibble(state = state.abb, state_name = state.name)
state_list

The state_map dataframe is in lower case and has the column name ‘region’. I shall change the state_list tibble to be the same format so they can be joined together.

state_list <- tibble(state = state.abb, region = tolower(state.name))

Joing state_list to fires_small datasets

fires_states <- fires_small %>%
  left_join(state_list, by = "state")

fires_states

Checking the join has worked and there are no missing values.

fires_states %>%
  filter(is.na(region))

There does seem to be 22,147 NAs in the ‘region’ column we just made. Scrolling through there are 2 missing States of ‘PR’ and ‘DC’ in the states_list tibble.

After some quick research it seems that there are only 50 States in the US. Washington DC is techincally not counted as a state but as a Federal District, as it is the seat of government, so that was why it wasn’t included in the States tibble originally. PR is Puerto Rico and is also not a state but the largest US territory .

I shall add DC and PR into the state_list and re-join it.

# Adding 2 new states

state.abb <- append(state.abb, c("DC", "PR"))
state.name <- append(state.name, c("District of Columbia", "Puerto Rico"))

state_list <- tibble(state = state.abb, region = tolower(state.name))

# Re-joing tibbles

fires_states <- fires_small %>%
  left_join(state_list, by = "state")

# Checking the join has worked properly and there are no NAs

fires_states %>%
  filter(is.na(region))

Warning in `[<-.data.frame`(`*tmp*`, is_list, value = list(`23` = "<S3: blob>")) :
  replacement element 1 has 1 row to replace 0 rows

# Code below brings up a "vector memory exhausted (limit reached?)" error

# fires_joined <- fires_states %>%
#  right_join(state_map, by = "region")

The data set and geo information is too big to join so I’m going to do a summarise first to get the number of fires per region first.

fires_joined <- fires_states %>% 
    select(region) %>%
    group_by(region) %>%
    summarise(num_fires = n()) %>%
    right_join(state_map, by = "region")

`summarise()` ungrouping output (override with `.groups` argument)

Result!! Now doing first geo spatial visualisation

Total Wildfires per state from 1992-2015

fires_joined %>% 
    ggplot +
    (aes(x = long, y = lat, group = group, fill = num_fires)) + 
    geom_polygon() + 
    geom_path(color = "white") + 
    scale_fill_continuous(low = "darkblue", 
                          high = "darkred",
                          name = "Number of fires") + 
    theme_map() + 
    coord_map("mollweide") + 
    ggtitle("Total US Wildfires from 1992-2015") + 
    theme(plot.title = element_text(hjust = 0.5))

3. Geo Spatial Visualisations

The dataset has a cause of fire column so I shall now create some causation plots.

Getting list of fire causes

fires_states %>%
  distinct(stat_cause_descr) %>%
  arrange(-desc(stat_cause_descr))

Total fire by cause in tabular form

fires_states %>%
  select(stat_cause_descr) %>%
  group_by(stat_cause_descr) %>%
  summarise(num_fires = n ()) %>%
  arrange(desc(num_fires))

`summarise()` ungrouping output (override with `.groups` argument)

NA

Number of fires by state in tabular form

fires_states %>%
  select(region) %>%
  group_by(region) %>%
  summarise(num_fires = n()) %>%
  arrange(desc(num_fires))

`summarise()` ungrouping output (override with `.groups` argument)

As the cause needs to be filtered before the map join, I’m going to either going to have to repeat a whole load of the same code in every single plot or write a function that will do it for me with, saving a lot of typing!

# Function for plotting cause of fire

cause <- function(cause) {
  fires_states %>%
    filter(stat_cause_descr == cause) %>%
    select(region) %>%
    group_by(region) %>%
    summarise(num_fires = n ()) %>%
    right_join(state_map, by = "region") %>%
    ggplot +
    (aes(x = long, y = lat, group = group, fill = num_fires)) + 
    geom_polygon() + 
    geom_path(color = "white") + 
    scale_fill_continuous(low = "darkblue", 
                          high = "darkred",
                          name = "Number of fires") + 
    theme_map() + 
    coord_map("mollweide") + 
    ggtitle(paste0("Total US Wildfires caused by ", cause, " from 1992-2015")) + 
    theme(plot.title = element_text(hjust = 0.5))
}

Wildfires caused by Arson

cause("Arson")

`summarise()` ungrouping output (override with `.groups` argument)

Arson does seem more prevalent in the SE states of Mississippi, Georgia, Alabama and also the western state of California.

Wildfires caused by Campfire

cause("Campfire")

`summarise()` ungrouping output (override with `.groups` argument)

Campfires are the most prevalent in the Western states of Oregon, California and Arizona.

Wildfires caused by Children

cause("Children")

`summarise()` ungrouping output (override with `.groups` argument)

Fires by children are spread about the country, but the most prevalent states are California in the West, Alabama and South Carolina and New Jersey in the east.

Wildfires caused by Debris Burning

cause("Debris Burning")

`summarise()` ungrouping output (override with `.groups` argument)

Fires by burning debris are mostly in the southern warmer states of Texas, Georgia and North Carolina.

Wildfires caused by Equiment Use

cause("Equipment Use")

`summarise()` ungrouping output (override with `.groups` argument)

Most of the fires caused by equipment seem to be in California

Wildfires caused by Fireworks

cause("Fireworks")

`summarise()` ungrouping output (override with `.groups` argument)

Most of the fires caused by fireworks seem to be in the north of the country. Primarily South Dakota, Montana and Washington state.

Wildfires caused by Lightning

cause("Lightning")

`summarise()` ungrouping output (override with `.groups` argument)

Apart from a hotspot of lightning strikes in Florida, the vast majority of fires caused by lightning are in the West of the country. With the 3 most affected states being California, Oregon and Arizona.

Wildfires caused by Miscellious

cause("Miscellaneous")

`summarise()` ungrouping output (override with `.groups` argument)

There seems to be quite a few miscellaneous classifications in California, Texas and New York.

Wildfires caused by Missing/Undefined

cause("Missing/Undefined")

`summarise()` ungrouping output (override with `.groups` argument)

The states with the most missing or undefined data is North and South Carolina, Oklahoma and California.

Wildfires caused by Powerline

cause("Powerline")

`summarise()` ungrouping output (override with `.groups` argument)

Texas has the largest amount of wildfires caused by powerlines. This is likely due to the warm climate and the large proportion of the state that is dry grasslands used for agriculture. (1)

1. https://uk.reuters.com/article/us-wildfires-texas/trees-and-power-lines-caused-major-texas-fire-idUSTRE78J76A20110920

Wildfires caused by Railroad

cause("Railroad")

`summarise()` ungrouping output (override with `.groups` argument)

By far Florida has the most wildfires caused by railroads.

Wildfires caused by Smoking

cause("Smoking")

`summarise()` ungrouping output (override with `.groups` argument)

Fires caused by smoking seem to be spread around the country, but mainly on the east and west coasts.

Wildfires caused by Structure

cause("Structure")

`summarise()` ungrouping output (override with `.groups` argument)

South Dakota has the largest proportion of fires caused by structures.

Unsurprisingly the southern states seem to have more occurences of wildifre in general, no doubt due to the warmer climate at their latitudes. Also the 1st and 3rd states with the highest number of fires are also the 2 largest States by size. However the 2nd highest State is Georgia, which although it is in the South of the country is only an average sized State. Therefore to get a better picture of what is going on I’m going to look at the proportion of fires occuring by square mile by normalising the State size.

The dataset package also has the area in square miles of each state included in the state.area vector.

state.area

 [1]  51609 589757 113909  53104 158693 104247   5009   2057  58560  58876
[11]   6450  83557  56400  36291  56290  82264  40395  48523  33215  10577
[21]   8257  58216  84068  47716  69686 147138  77227 110540   9304   7836
[31] 121666  49576  52586  70665  41222  69919  96981  45333   1214  31055
[41]  77047  42244 267339  84916   9609  40815  68192  24181  56154  97914

length(state.area)

[1] 50

Annoyingly it also only has 50 states not 52 so I will need to add in DC and PR back in.

(Area figures obtained from Wikipedia)

DC = 68 miles^2 PR = 3515 miles^2

# To make my life easier I'm going to remove the state.abb and .name files and make the tibble again, adding in the land area figures at the same time to make sure they are in the correct order.

rm(state.abb)
rm(state.name)

state.abb <- append(state.abb, c("DC", "PR"))
state.name <- append(state.name, c("District of Columbia", "Puerto Rico"))
state.area <- append(state.area, c("68", "3515"))

state_list <- tibble(state = state.abb, region = tolower(state.name), area = as.numeric(state.area))

# Re-joining tibbles

fires_states <- fires_small %>%
  left_join(state_list, by = "state")

Normalising States area sizes

fires_states %>%
  select(region, area) %>%
  group_by(region, area) %>%
  summarise(num_fires = n()) %>%
  mutate(fires_sqmile = num_fires / area) %>%
  arrange(desc(fires_sqmile))

`summarise()` regrouping output by 'region' (override with `.groups` argument)

This table shows Puerto Rico has the highest proportion of fires compared to its size, followed by New Jersey in the NE of the country and finally by the States in the SE of the country.

fires_states %>%
  select(region, area) %>%
  group_by(region, area) %>%
  summarise(num_fires = n()) %>%
  mutate(fires_sqmile = num_fires / area) %>%
  right_join(state_map, by = "region") %>%
  ggplot +
  (aes(x = long, y = lat, group = group, fill = fires_sqmile)) + 
  geom_polygon() + 
  geom_path(color = "white") + 
  scale_fill_distiller(name = "Fire per Sq Mile", palette = "PuBuGn") +
  theme_map() + 
  coord_map("mollweide") + 
  ggtitle(paste0("Total US Wildfires per Square Mile from 1992-2015")) + 
  theme(plot.title = element_text(hjust = 0.5))

`summarise()` regrouping output by 'region' (override with `.groups` argument)

Puerto Rico is not shown on this map, but visually we can see the data for the other 51 entries, and the south eastern states still have the highest proportion of wildfires. Interestingly New Jersey also shows has a hotspot in the NE of the country.

Do causes change over time?

Splitting causes into 2 group for legibility.

The first group is for directly man created fires.

fires_states %>%
  select(stat_cause_descr, fire_year) %>%
  group_by(fire_year, stat_cause_descr) %>%
  filter(stat_cause_descr == "Arson" | stat_cause_descr == "Campfire" |
           stat_cause_descr == "Children" | stat_cause_descr == "Equipment Use" |
           stat_cause_descr == "Fireworks" | stat_cause_descr == "Smoking") %>%
  summarise(num_fires = n()) %>%
  ggplot +
  aes(x = fire_year, y = num_fires, colour = stat_cause_descr) +
  geom_line()

`summarise()` regrouping output by 'fire_year' (override with `.groups` argument)

The 2 large peaks in Arson are obvious in 1999 and 2006. There was a large heatwave in 2006, but I’m not sure why this would result in an increase in arson. Unless this was just due to the dry ground creating extra fuel to aid the spread of fires that would have normally not resulted in a large scale fire. This may also be the same reason that there is also another peak in 2006 for Equipment Use. Arson however does look to be decreasing since 2006.

And this one for natural occuring fires.

fires_states %>%
  select(stat_cause_descr, fire_year) %>%
  group_by(fire_year, stat_cause_descr) %>%
  filter(stat_cause_descr == "Debris Burning" | stat_cause_descr == "Lightning" |
           stat_cause_descr == "Miscellaneous" | stat_cause_descr == 
           "Missing/Undefined" | stat_cause_descr == "Powerline" | 
           stat_cause_descr == "Railroad" | stat_cause_descr == "Structure") %>%
  summarise(num_fires = n()) %>%
  ggplot +
  aes(x = fire_year, y = num_fires, colour = stat_cause_descr) +
  geom_line()

`summarise()` regrouping output by 'fire_year' (override with `.groups` argument)

Similar peaks can be seen in Debris, Miscellaneous and lightning in the heatwave of 2006 that left the ground very dry. There are peaks from 1997 to 2003 in debris, miscellaneous and lightening, but also a trough in missing/undefined, so this is likely to be due to more accurate classification of fires and not using the missing/undefined category as much.

Difference in causes between states

state_map_southern <- state_map %>%
  filter(region == "florida" | region == "georgia" | region == "alabama" |
           region == "mississippi" | region == "south carolina" | 
           region == "north carolina" | region == "tennessee" | 
           region == "arkansas" | region == "louisiana")

fires_states %>%
  filter(fire_year == "1992" | fire_year == "1993" | fire_year == "1994" |
           fire_year == "1995") %>%
  filter(region == "florida" | region == "georgia" | region == "alabama" |
           region == "mississippi" | region == "south carolina" | 
           region == "north carolina" | region == "tennessee" |
           region == "arkansas" | region == "louisiana") %>%
  select(region, stat_cause_descr) %>%
  group_by(region, stat_cause_descr) %>%
  summarise(num_fire = n()) %>%
  top_n(1) %>%
  right_join(state_map_southern, by = "region") %>%
  ggplot +
  (aes(x = long, y = lat, group = group, fill = stat_cause_descr)) + 
  geom_polygon() + 
  geom_path(color = "white") + 
  theme_map() + 
  scale_fill_brewer(name = "Cause of Fires", palette = "PuBuGn") +
  ggtitle("Total US Wildfires main cause from 1992-1995") + 
  theme(plot.title = element_text(hjust = 0.5))

`summarise()` regrouping output by 'region' (override with `.groups` argument)
Selecting by num_fire

fires_states %>%
  filter(fire_year == "1996" | fire_year == "1997" | fire_year == "1998" |
           fire_year == "1999") %>%
  filter(region == "florida" | region == "georgia" | region == "alabama" |
           region == "mississippi" | region == "south carolina" | 
           region == "north carolina" | region == "tennessee" |
           region == "arkansas" | region == "louisiana") %>%
  select(region, stat_cause_descr) %>%
  group_by(region, stat_cause_descr) %>%
  summarise(num_fire = n()) %>%
  top_n(1) %>%
  right_join(state_map_southern, by = "region") %>%
  ggplot +
  (aes(x = long, y = lat, group = group, fill = stat_cause_descr)) + 
  geom_polygon() + 
  geom_path(color = "white") + 
  theme_map() + 
  scale_fill_brewer(name = "Cause of Fires", palette = "PuBuGn") +
  ggtitle("Total US Wildfires main cause from 1996-1999") + 
  theme(plot.title = element_text(hjust = 0.5))

`summarise()` regrouping output by 'region' (override with `.groups` argument)
Selecting by num_fire

fires_states %>%
  filter(fire_year == "2000" | fire_year == "2001" | fire_year == "2002" |
           fire_year == "2003") %>%
  filter(region == "florida" | region == "georgia" | region == "alabama" |
           region == "mississippi" | region == "south carolina" | 
           region == "north carolina" | region == "tennessee" |
           region == "arkansas" | region == "louisiana") %>%
  select(region, stat_cause_descr) %>%
  group_by(region, stat_cause_descr) %>%
  summarise(num_fire = n()) %>%
  top_n(1) %>%
  right_join(state_map_southern, by = "region") %>%
  ggplot +
  (aes(x = long, y = lat, group = group, fill = stat_cause_descr)) + 
  geom_polygon() + 
  geom_path(color = "white") + 
  theme_map() + 
  scale_fill_brewer(name = "Cause of Fires", palette = "PuBuGn") +
  ggtitle("Total US Wildfires main cause from 2000-2003") + 
  theme(plot.title = element_text(hjust = 0.5))

`summarise()` regrouping output by 'region' (override with `.groups` argument)
Selecting by num_fire

fires_states %>%
  filter(fire_year == "2004" | fire_year == "2005" | fire_year == "2006" |
           fire_year == "2007") %>%
  filter(region == "florida" | region == "georgia" | region == "alabama" |
           region == "mississippi" | region == "south carolina" | 
           region == "north carolina" | region == "tennessee" |
           region == "arkansas" | region == "louisiana") %>%
  select(region, stat_cause_descr) %>%
  group_by(region, stat_cause_descr) %>%
  summarise(num_fire = n()) %>%
  top_n(1) %>%
  right_join(state_map_southern, by = "region") %>%
  ggplot +
  (aes(x = long, y = lat, group = group, fill = stat_cause_descr)) + 
  geom_polygon() + 
  geom_path(color = "white") + 
  theme_map() + 
  scale_fill_brewer(name = "Cause of Fires", palette = "PuBuGn") +
  ggtitle("Total US Wildfires main cause from 2004-2007") + 
  theme(plot.title = element_text(hjust = 0.5))

`summarise()` regrouping output by 'region' (override with `.groups` argument)
Selecting by num_fire

fires_states %>%
  filter(fire_year == "2008" | fire_year == "2009" | fire_year == "2010" |
           fire_year == "2011") %>%
  filter(region == "florida" | region == "georgia" | region == "alabama" |
           region == "mississippi" | region == "south carolina" | 
           region == "north carolina" | region == "tennessee" |
           region == "arkansas" | region == "louisiana") %>%
  select(region, stat_cause_descr) %>%
  group_by(region, stat_cause_descr) %>%
  summarise(num_fire = n()) %>%
  top_n(1) %>%
  right_join(state_map_southern, by = "region") %>%
  ggplot +
  (aes(x = long, y = lat, group = group, fill = stat_cause_descr)) + 
  geom_polygon() + 
  geom_path(color = "white") + 
  theme_map() + 
  scale_fill_brewer(name = "Cause of Fires", palette = "PuBuGn") +
  ggtitle("Total US Wildfires main cause from 2008-2011") + 
  theme(plot.title = element_text(hjust = 0.5))

`summarise()` regrouping output by 'region' (override with `.groups` argument)
Selecting by num_fire

fires_states %>%
  filter(fire_year == "2012" | fire_year == "2013" | fire_year == "2014" |
           fire_year == "2015") %>%
  filter(region == "florida" | region == "georgia" | region == "alabama" |
           region == "mississippi" | region == "south carolina" | 
           region == "north carolina" | region == "tennessee" |
           region == "arkansas" | region == "louisiana") %>%
  select(region, stat_cause_descr) %>%
  group_by(region, stat_cause_descr) %>%
  summarise(num_fire = n()) %>%
  top_n(1) %>%
  right_join(state_map_southern, by = "region") %>%
  ggplot +
  (aes(x = long, y = lat, group = group, fill = stat_cause_descr)) + 
  geom_polygon() + 
  geom_path(color = "white") + 
  theme_map() + 
  scale_fill_brewer(name = "Cause of Fires", palette = "PuBuGn") +
  ggtitle("Total US Wildfires main cause from 2012-2015") + 
  theme(plot.title = element_text(hjust = 0.5))

`summarise()` regrouping output by 'region' (override with `.groups` argument)
Selecting by num_fire

Looking at these trends some interesting insights can be seen. For the combined years data Florida stands out as having railroad as its main cause of wildfire, but from the above plots it can be seen that these railroad fires are only the main cause up to the 4 yearly period ending in 2003 and then the main cause changes to lightning until the end of the collection period in 2015. Similarly arson seem reasonably popular in the southern states until 2007, when it no longer appears as the most common cause of wildfire. This downward trend was also noted earlier in the overall causation plots for all states

Correlation between states and fire size

fires_states %>%
  select(region, fire_size_class) %>%
  group_by(region, fire_size_class) %>%
  summarise(num_fire = n()) %>%
  top_n(1) %>%
  right_join(state_map, by = "region") %>%
  ggplot +
  (aes(x = long, y = lat, group = group, fill = fire_size_class)) +
  geom_polygon() +
  geom_path(color = "white") +
  theme_map() +
  scale_fill_brewer(name = "Fire Size Class", palette = "PuBuGn") +
  ggtitle("Most common wildfire size per State 1992-2015") +
  theme(plot.title = element_text(hjust = 0.5))

`summarise()` regrouping output by 'region' (override with `.groups` argument)
Selecting by num_fire

fires_states %>%
  select(region, fire_size_class) %>%
  filter(fire_size_class == "G") %>%
  group_by(region) %>%
  summarise(num_fire = n()) %>%
  right_join(state_map, by = "region") %>%
  ggplot +
  (aes(x = long, y = lat, group = group, fill = num_fire)) +
  geom_polygon() +
  geom_path(color = "white") +
  theme_map() +
  scale_fill_distiller(name = "Number of Fires", palette = "PuBuGn") +
  ggtitle("Number of large class G fires per State 1992-2015") +
  theme(plot.title = element_text(hjust = 0.5))

`summarise()` ungrouping output (override with `.groups` argument)

From the plots we can see that the Western states have the most small fires and also the most large fires! Not entirely the most helpful plots…

Are fires more prevalent in certain months for individual states

fires_states %>%
  filter(fire_year == "1992" | fire_year == "1993" | fire_year == "1994" |
           fire_year == "1995") %>%
  select(region, discovery_moy) %>%
  group_by(region, discovery_moy) %>%
  summarise(num_fire = n()) %>%
  top_n(1) %>%
  right_join(state_map, by = "region") %>%
  ggplot +
  (aes(x = long, y = lat, group = group, fill = discovery_moy)) +
  geom_polygon() +
  geom_path(color = "white") +
  theme_map() +
  scale_fill_brewer(name = "Months of Year", palette = "PuBuGn") +
  ggtitle("Month with most fires per State 1992-1995") +
  theme(plot.title = element_text(hjust = 0.5))

`summarise()` regrouping output by 'region' (override with `.groups` argument)
Selecting by num_fire

fires_states %>%
  filter(fire_year == "1996" | fire_year == "1997" | fire_year == "1998" |
           fire_year == "1999") %>%
  select(region, discovery_moy) %>%
  group_by(region, discovery_moy) %>%
  summarise(num_fire = n()) %>%
  top_n(1) %>%
  right_join(state_map, by = "region") %>%
  ggplot +
  (aes(x = long, y = lat, group = group, fill = discovery_moy)) +
  geom_polygon() +
  geom_path(color = "white") +
  theme_map() +
  scale_fill_brewer(name = "Months of Year", palette = "PuBuGn") +
  ggtitle("Month with most fires per State 1996-1999") +
  theme(plot.title = element_text(hjust = 0.5))

`summarise()` regrouping output by 'region' (override with `.groups` argument)
Selecting by num_fire

fires_states %>%
  filter(fire_year == "2000" | fire_year == "2001" | fire_year == "2002" |
           fire_year == "2003") %>%
  select(region, discovery_moy) %>%
  group_by(region, discovery_moy) %>%
  summarise(num_fire = n()) %>%
  top_n(1) %>%
  right_join(state_map, by = "region") %>%
  ggplot +
  (aes(x = long, y = lat, group = group, fill = discovery_moy)) +
  geom_polygon() +
  geom_path(color = "white") +
  theme_map() +
  scale_fill_brewer(name = "Months of Year", palette = "PuBuGn") +
  ggtitle("Month with most fires per State 2000-2003") +
  theme(plot.title = element_text(hjust = 0.5))

`summarise()` regrouping output by 'region' (override with `.groups` argument)
Selecting by num_fire

fires_states %>%
  filter(fire_year == "2004" | fire_year == "2005" | fire_year == "2006" |
           fire_year == "2007") %>%
  select(region, discovery_moy) %>%
  group_by(region, discovery_moy) %>%
  summarise(num_fire = n()) %>%
  top_n(1) %>%
  right_join(state_map, by = "region") %>%
  ggplot +
  (aes(x = long, y = lat, group = group, fill = discovery_moy)) +
  geom_polygon() +
  geom_path(color = "white") +
  theme_map() +
  scale_fill_brewer(name = "Months of Year", palette = "PuBuGn") +
  ggtitle("Month with most fires per State 2004-2007") +
  theme(plot.title = element_text(hjust = 0.5))

`summarise()` regrouping output by 'region' (override with `.groups` argument)
Selecting by num_fire

fires_states %>%
  filter(fire_year == "2008" | fire_year == "2009" | fire_year == "2010" |
           fire_year == "2011") %>%
  select(region, discovery_moy) %>%
  group_by(region, discovery_moy) %>%
  summarise(num_fire = n()) %>%
  top_n(1) %>%
  right_join(state_map, by = "region") %>%
  ggplot +
  (aes(x = long, y = lat, group = group, fill = discovery_moy)) +
  geom_polygon() +
  geom_path(color = "white") +
  theme_map() +
  scale_fill_brewer(name = "Months of Year", palette = "PuBuGn") +
  ggtitle("Month with most fires per State 2008-2011") +
  theme(plot.title = element_text(hjust = 0.5))

`summarise()` regrouping output by 'region' (override with `.groups` argument)
Selecting by num_fire

fires_states %>%
  filter(fire_year == "2012" | fire_year == "2013" | fire_year == "2014" |
           fire_year == "2015") %>%
  select(region, discovery_moy) %>%
  group_by(region, discovery_moy) %>%
  summarise(num_fire = n()) %>%
  top_n(1) %>%
  right_join(state_map, by = "region") %>%
  ggplot +
  (aes(x = long, y = lat, group = group, fill = discovery_moy)) +
  geom_polygon() +
  geom_path(color = "white") +
  theme_map() +
  scale_fill_brewer(name = "Months of Year", palette = "PuBuGn") +
  ggtitle("Month with most fires per State 2012-2015") +
  theme(plot.title = element_text(hjust = 0.5))

`summarise()` regrouping output by 'region' (override with `.groups` argument)
Selecting by num_fire

fires_states %>%
  filter(fire_year == "2010" | fire_year == "2011" | fire_year == "2012" |
           fire_year == "2013" | fire_year == "2014" | fire_year == "2015") %>%
  select(region, discovery_moy) %>%
  group_by(region, discovery_moy) %>%
  summarise(num_fire = n()) %>%
  top_n(1) %>%
  right_join(state_map, by = "region") %>%
  ggplot +
  (aes(x = long, y = lat, group = group, fill = discovery_moy)) +
  geom_polygon() +
  geom_path(color = "white") +
  theme_map() +
  scale_fill_brewer(name = "Months of Year", palette = "PuBuGn") +
  ggtitle("Month with most fires in per State 2010-2015") +
  theme(plot.title = element_text(hjust = 0.5))

`summarise()` regrouping output by 'region' (override with `.groups` argument)
Selecting by num_fire

The above plots are quite interesting. The months of the year that have the most seems to widely change in certain state. Mainly the east half of the country have the most fires in the Spring (Feb-May) and the western part of the country have the most fires later on in Summer and Fall (Jun-Oct). There are however a few exceptions that can be seen in the 2004-2007 and 2008-2011 data Texas has the most fires in January. Florida also mostly conformed to the East/West split with the majority of its worst months for fires taking place in March or April up until 2007, then the most common month moves later into June and July for the rest of the reporting period until 2015. This may have to due with main cause of fires in Florida changing from railroad to lightning related about the same time, as we noted earlier on when looking at causation. As July is the main month for tropical storms and lightning in Florida this is a possible cause for the highest month becoming later in the year than before. (2)

2. https://www.weather.gov/mlb/fl_lightning_climo